Offline Reinforcement Learning from Vision-Language Model Feedback for Real-World Robot-Assisted Dressing and Simulated Tasks
This paper introduces Offline RL-VLM-F, a novel system that leverages vision-language models (VLMs) to automatically generate reward labels for unlabeled datasets, enabling offline reinforcement learning for complex real-world robotics tasks, such as robot-assisted dressing, and outperforming existing baselines in various simulated manipulation tasks.
Constrained Latent Action Policies (C-LAP): A Novel Approach to Model-Based Offline Reinforcement Learning for Improved Performance and Reduced Value Overestimation
C-LAP, a novel model-based offline reinforcement learning method, leverages a generative model of joint state-action distribution and a constrained policy optimization approach to enhance performance and mitigate value overestimation, particularly excelling in scenarios with visual observations.
Hypercube Policy Regularization: Enhancing Offline Reinforcement Learning by Exploring Similar State Actions
The hypercube policy regularization framework improves offline reinforcement learning by allowing agents to explore actions corresponding to similar states within a hypercube, striking a balance between conservatism and aggressiveness for better policy learning.
Domain Adaptation for Offline Reinforcement Learning with Limited Samples: A Theoretical and Empirical Exploration
This paper proposes a novel framework for domain adaptation in offline reinforcement learning (RL) with limited target samples, theoretically analyzing the trade-off between leveraging a large, related source dataset and a limited target dataset, and providing empirical validation on the Procgen benchmark.
Diffusion Trusted Q-Learning: A Dual Policy Approach for Efficient and Effective Offline Reinforcement Learning
Diffusion Trusted Q-Learning (DTQL) leverages the expressiveness of diffusion policies for behavior cloning while employing a novel diffusion trust region loss to guide a computationally efficient one-step policy for superior performance in offline reinforcement learning tasks.
Diffusion-DICE: Using Diffusion Models to Improve Offline Reinforcement Learning by Transforming Behavior Policy to Optimal Policy
Diffusion-DICE is a novel offline reinforcement learning algorithm that leverages diffusion models to transform the behavior policy distribution into an optimal policy distribution, achieving state-of-the-art performance by minimizing error exploitation in value function approximation.
Latent Plan Transformer (LPT): A Novel Approach to Planning in Offline Reinforcement Learning Using Trajectory-Return Data
LPT, a novel generative model, effectively performs planning in offline reinforcement learning by leveraging a latent variable to connect trajectory generation with final returns, achieving temporal consistency and outperforming existing methods in challenging tasks.
離線行為蒸餾：從次優數據中提取高效策略
本文提出了一種名為離線行為蒸餾（OBD）的新方法，旨在從大量的次優強化學習數據中提取精簡的專家行為數據，從而實現快速且高效的策略學習。
Decision Mamba: An Efficient Transformer Alternative for Trajectory Optimization in Offline Reinforcement Learning
Mamba, a linear-time sequence model, can be effectively adapted for trajectory optimization in offline reinforcement learning, achieving comparable or superior performance to transformer-based methods while using significantly fewer parameters.
Offline Reinforcement Learning in Combinatorial Action Spaces: Introducing Branch Value Estimation
Branch Value Estimation (BVE) is a novel offline reinforcement learning method that effectively addresses the challenges of learning in large, discrete combinatorial action spaces by representing the action space as a tree and learning to evaluate only a small subset of actions at each timestep.
© 2024 by Linnk AI